A GPU-accelerated adaptive kernel density estimation approach for efficient point pattern analysis on spatial big data

doi:10.1080/13658816.2017.1324975

	A GPU-accelerated adaptive kernel density estimation approach for efficient point pattern analysis on spatial big data
	Zhang, Guiming 1; Zhu, A-Xing 1,2,3,4,5; Huang, Qunying 1
刊名	INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE
	2017
卷号	31 期号:10 页码:2068-2097
关键词	Adaptive kernel density estimation optimization GPU/CUDA OpenMP spatial big data
ISSN号	1365-8816
DOI	10.1080/13658816.2017.1324975
通讯作者	Zhu, A-Xing(azhu@wisc.edu)
英文摘要	Kernel density estimation (KDE) is a classic approach for spatial point pattern analysis. In many applications, KDE with spatially adaptive bandwidths (adaptive KDE) is preferred over KDE with an invariant bandwidth (fixed KDE). However, bandwidths determination for adaptive KDE is extremely computationally intensive, particularly for point pattern analysis tasks of large problem sizes. This computational challenge impedes the application of adaptive KDE to analyze large point data sets, which are common in this big data era. This article presents a graphics processing units (GPUs)-accelerated adaptive KDE algorithm for efficient spatial point pattern analysis on spatial big data. First, optimizations were designed to reduce the algorithmic complexity of the bandwidth determination algorithm for adaptive KDE. The massively parallel computing resources on GPU were then exploited to further speed up the optimized algorithm. Experimental results demonstrated that the proposed optimizations effectively improved the performance by a factor of tens. Compared to the sequential algorithm and an Open Multiprocessing (OpenMP)-based algorithm lever-aging multiple central processing unit cores for adaptive KDE, the GPU-enabled algorithm accelerated point pattern analysis tasks by a factor of hundreds and tens, respectively. Additionally, the GPU-accelerated adaptive KDE algorithm scales reasonably well while increasing the size of data sets. Given the significant acceleration brought by the GPU-enabled adaptive KDE algorithm, point pattern analysis with the adaptive KDE approach on large point data sets can be performed efficiently. Point pattern analysis on spatial big data, computationally prohibitive with the sequential algorithm, can be conducted routinely with the GPU-accelerated algorithm. The GPU-accelerated adaptive KDE approach contributes to the geospatial computational toolbox that facilitates geographic knowledge discovery from spatial big data.
资助项目	National Natural Science Foundation of China[41431177] ; National Basic Research Program of China[2015CB954102] ; Natural Science Research Program of Jiangsu[14KJA170001] ; PAPD ; National Key Technology Innovation Project for Water Pollution Control and Remediation[2013ZX07103006] ; Vilas Associate Award ; Hammel Faculty Fellow Award ; Manasse Chair Professorship from the University of Wisconsin-Madison ; 'One-Thousand Talents' Program of China
WOS关键词	BANDWIDTH ; CYBERINFRASTRUCTURE ; GEOCOMPUTATION ; ALGORITHM ; SELECTION ; SCIENCES ; NETWORK ; RANGE ; MODEL ; CUDA
WOS研究方向	Computer Science ; Geography ; Physical Geography ; Information Science & Library Science
语种	英语
出版者	TAYLOR & FRANCIS LTD
WOS记录号	WOS:000405678300008
资助机构	National Natural Science Foundation of China ; National Basic Research Program of China ; Natural Science Research Program of Jiangsu ; PAPD ; National Key Technology Innovation Project for Water Pollution Control and Remediation ; Vilas Associate Award ; Hammel Faculty Fellow Award ; Manasse Chair Professorship from the University of Wisconsin-Madison ; 'One-Thousand Talents' Program of China
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/62782]
专题	中国科学院地理科学与资源研究所
通讯作者	Zhu, A-Xing
作者单位	1.Univ Wisconsin, Dept Geog, Madison, WI 53706 USA 2.Nanjing Normal Univ, Minist Educ, Key Lab Virtual Geog Environm, Nanjing, Jiangsu, Peoples R China 3.State Key Lab Cultivat Base Geog Environm Evolut, Nanjing, Jiangsu, Peoples R China 4.Chinese Acad Sci, State Key Lab Resources & Environm Informat Syst, Inst Geog Sci & Nat Resources Res, Beijing, Peoples R China 5.Jiangsu Ctr Collaborat Innovat Geog Informat Res, Nanjing, Jiangsu, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Guiming,Zhu, A-Xing,Huang, Qunying. A GPU-accelerated adaptive kernel density estimation approach for efficient point pattern analysis on spatial big data[J]. INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE,2017,31(10):2068-2097.
APA	Zhang, Guiming,Zhu, A-Xing,&Huang, Qunying.(2017).A GPU-accelerated adaptive kernel density estimation approach for efficient point pattern analysis on spatial big data.INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE,31(10),2068-2097.
MLA	Zhang, Guiming,et al."A GPU-accelerated adaptive kernel density estimation approach for efficient point pattern analysis on spatial big data".INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE 31.10(2017):2068-2097.